AI Roadmaps
← All roadmaps
AdvancedRoadmap 41 · 12 steps · 4h 52m of video

Self-Supervised & Contrastive Learning

How models learn strong features from unlabeled data: contrastive losses, SimCLR, BYOL, masked autoencoders, DINO through DINOv3, and JEPA's predict-the-representation idea.

Before this: Deep Learning and Computer Vision, including vision transformers. Multimodal & Vision-Language Models covers CLIP, the image-text side of contrastive learning.

  1. 10 min
    Step 1

    What self-supervised learning is

    Learn features from unlabeled data with training targets made from the data itself.

    What Is Self-Supervised Learning and Why Care?

    Boris Meinardus

  2. 1h 15m
    Step 2

    Self-supervised learning in CS231N

    Stanford lecture tour from pretext tasks to contrastive and masked methods.

    Stanford CS231N | Spring 2025 | Lecture 12: Self-Supervised Learning

    Stanford Online

  3. 5 min
    Step 3

    Contrastive learning in five minutes

    The core idea: similar pairs close together, dissimilar pairs far apart.

    Contrastive Learning - 5 Minutes with Cyrill

    Cyrill Stachniss

  4. 20 min
    Step 4

    Contrastive loss

    The loss behind that idea, worked through step by step.

    Contrastive Loss : Data Science Basics

    ritvikmath

  5. 15 min
    Step 5

    SimCLR

    Two augmented views of one image should match; other batch images are negatives.

    Contrastive Learning with SimCLR | Deep Learning Animated

    Deepia

    sthalles/SimCLR★ 2.5k
  6. 34 min
    Step 6

    BYOL: no negative pairs

    An online network predicts a slow momentum target's embedding, without negatives.

    BYOL: Bootstrap Your Own Latent: A New Approach to Self-Supervised Learning (Paper Explained)

    Yannic Kilcher

    lucidrains/byol-pytorch★ 1.9k
  7. 13 min
    Step 7

    Masked autoencoders

    Mask most image patches and train a ViT to reconstruct the missing pixels.

    Masked Autoencoders Are Scalable Vision Learners – Paper explained and animated!

    AI Coffee Break with Letitia

  8. 39 min
    Step 8

    DINO: self-distillation

    A student matches a momentum teacher across image crops, with no labels.

    DINO: Emerging Properties in Self-Supervised Vision Transformers (Facebook AI Research Explained)

    Yannic Kilcher

  9. 8 min
    Step 9

    DINOv2

    DINO-style training scaled on curated data, giving general-purpose frozen features.

    DINOv2 from Meta AI - Finally a Foundational Model in Computer Vision?

    AI Papers Academy

    facebookresearch/dinov2★ 13.6k
  10. 16 min
    Step 10

    DINOv3

    Larger-scale training that keeps dense patch features useful for many tasks.

    How AI Taught Itself to See [DINOv3]

    Jia-Bin Huang

    facebookresearch/dinov3★ 11.8k
  11. 8 min
    Step 11

    I-JEPA

    Predict the representations of masked image regions instead of their pixels.

    I-JEPA from Meta AI - A Human-Like Computer Vision Model | Paper Summary

    AI Papers Academy

  12. 50 min
    Step 12

    V-JEPA

    The same feature-prediction idea applied to unlabeled video clips.

    V-JEPA: Revisiting Feature Prediction for Learning Visual Representations from Video (Explained)

    Yannic Kilcher

    facebookresearch/jepa★ 4.2k

Build it

Watching isn't learning. Pick one and put it on GitHub.

Next roadmap